回归问题

最新推荐文章于 2024-02-18 23:02:05 发布

初学者___

最新推荐文章于 2024-02-18 23:02:05 发布

阅读量288

点赞数

文章标签： tensorflow

本文链接：https://blog.csdn.net/weixin_49305355/article/details/115800484

版权

学习笔记（1）

一、解析法实现一元线性回归

一元线性模型
参数求解

代码如下（示例）：

# 解析法实现一元线性回归

# 手动导入数据
x = [137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
     106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21]
y = [145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
     62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30]

# 均值
meanX = sum(x) / len(x)
meanY = sum(y) / len(y)

# 计算w,b
sumXY = 0.0
sumX = 0.0
for i in range(len(x)):
    sumXY += (x[i] - meanX) * (y[i] - meanY)
    sumX += (x[i] - meanX) * (x[i] - meanX)
w = sumXY / sumX
b = meanY - w*meanX
print("w=", w)
print("b=", b)

# 预测房价
x_test = [128.15, 45.00, 141.43, 106.27, 99.00, 53.84, 85.36, 70.00]
print("面积\t估计房价")
for i in range(len(x_test)):
    print(x_test[i], "\t", round(w*x_test[i]+b,2))

# Numpy实现
import numpy as np
from six import byte2int

x1 = np.array([137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
              106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21])
y1 = np.array([145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
              62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30])
meanX1 = np.mean(x1)
meanY1 = np.mean(y1)
sumXY1 = np.sum((x1-meanX1) * (y1-meanY1))
sumX1 = np.sum((x1-meanX1) * (x1-meanX1))
w1 = sumXY1 / sumX1
b1 = meanY1 - w1*meanX1
print("w1=", w1)
print("b1=", b1)
x_test1 = np.array([128.15, 45.00, 141.43, 106.27, 99.00, 53.84, 85.36, 70.00])
y_pred1 = w1*x_test1+b1
print("面积\t估计房价")
for i in range(len(x_test1)):
    print(x_test[i], "\t", round(y_pred1[i], 2))

# TensorFlow实现
import tensorflow as tf
x2 = tf.constant([137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
                  106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21])
y2 = tf.constant([145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
                  62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30])
meanX2 = tf.reduce_mean(x2)
meanY2 = tf.reduce_mean(y2)
sumXY2 = tf.reduce_sum((x2-meanX2) * (y2-meanY2))
sumX2 = tf.reduce_sum((x2-meanX2) * (x2-meanX2))
w2 = sumXY2 / sumX2
b2 = meanY2 - w2*meanX2
print("w2=", w2)
print("b2=", byte2int)
x_test2 = tf.constant([128.15, 45.00, 141.43, 106.27, 99.00, 53.84, 85.36, 70.00])
y_pred2 = w2*x_test2+b2
print("面积\t估计房价")
for i in range(len(x_test2)):
    print(x_test2[i], "\t", y_pred2[i])

完整代码：

# 解析法实现一元线性回归(完整程序)

import numpy as np
import tensorflow as tf
import matplotlib.pyplot as plt

with tf.device('/gpu:0'):
    # 设置字体
    plt.rcParams['font.sans-serif'] = ['SimHei']

    # 加载样本数据
    x = tf.constant([137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
                    106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21])
    y = tf.constant([145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
                    62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30])

    # 学习模型
    meanX = tf.reduce_mean(x)
    meanY = tf.reduce_mean(y)
    sumXY = tf.reduce_sum((x-meanX) * (y-meanY))
    sumX = tf.reduce_sum((x-meanX) * (x-meanX))
    w = sumXY / sumX
    b = meanY - w*meanX
    print("权值w=", w.numpy(),"\n偏置值b=", b.numpy())
    print("线性模型:y=", w.numpy(), "*x+", b.numpy())
    x_test = np.array([128.15, 45.00, 141.43, 106.27, 99.00, 53.84, 85.36, 70.00])
    y_pred = (w*x_test+b).numpy()

    # 预测房价
    print("面积\t估计房价")
    n = len(x_test)
    for i in range(n):
        print(x_test[i], "\t", round(y_pred[i], 2))

    # 数据和模型可视化
    plt.figure()

    plt.scatter(x,y,color="red",label="销售记录")
    plt.scatter(x_test,y_pred,color="blue",label="预测房价")
    plt.plot(x_test,y_pred,color="green",label="拟合直线",linewidth=2)

    plt.xlabel("面积（平方米）", fontsize=14)
    plt.ylabel("价格（万元）", fontsize=14)

    plt.xlim((40,150))
    plt.ylim((40,150))

    plt.suptitle("商品房销售价格评估系统v1.0", fontsize=20)

    plt.legend(loc="upper left")
    plt.show()

预测结果
在这里插入图片描述

二、解析法实现多元线性回归

多元线性回归

求解过程

代码如下（示例）：

# 解析法实现多元线性回归

#加载样本数据
import numpy as np

x1 = np.array([137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
               106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21])
x2 = np.array([3,2,2,3,1,2,3,2,2,3,1,1,1,1,2,2])
y = np.array([145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
              62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30])

# 数据处理
x0 = np.ones(len(x1))
X = np.stack((x0,x1,x2), axis = 1)
Y = np.array(y).reshape(-1,1)

# 求解模型参数
Xt = np.transpose(X)                    # 计算X'
XtX_1 = np.linalg.inv(np.matmul(Xt,X))  # 计算(X'X)-1
XtX_1_Xt = np.matmul(XtX_1, Xt)         # 计算(X'X-1)X'
W = np.matmul(XtX_1_Xt, Y)              # W=(X'X-1)X'Y

W = W.reshape(-1)
print("多元线性回归方程：")
print("Y=", W[1], "*x1+", W[2], "*x2+", W[0])

# 预测房价
print("请输入房屋面积和房间数，预测房屋销售价格：")
x1_test = float(input("商品房面积："))
x2_test = int(input("房间数："))
y_pred = W[1]*x1_test + W[2]*x2_test + W[0]
print("预测价格：", round(y_pred,2),"万元")

完整代码：

# 解析法实现多元线性回归(tensorflow)

#加载样本数据
import tensorflow as tf
import numpy as np

x1 = tf.constant([137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
               106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21])
x2 = tf.constant([3,2,2,3,1,2,3,2,2,3,1,1,1,1,2,2])
y = tf.constant([145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
              62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30])

# 数据处理
x0 = tf.ones(len(x1))
x2 = tf.cast(x2, dtype=tf.float32)
X = tf.stack((x0,x1,x2), axis = 1)
Y = tf.reshape(y,[-1,1])
#Y = tf.constant((y.numpy().reshape(-1,1)))

# 求解模型参数
Xt = tf.transpose(X, perm=[1,0])    # 计算X'
XtX = Xt@X
XtX_1 = tf.linalg.inv(XtX)     # 计算X'X-1
W = XtX_1@Xt@Y                      # W=(X'X-1)X'Y

W = tf.reshape(W, -1)
W = W.numpy()
print("多元线性回归方程：")
print("Y=", W[1], "*x1+", W[2], "*x2+", W[0])

# 预测房价
print("请输入房屋面积和房间数，预测房屋销售价格：")
x1_test = float(input("商品房面积："))
x2_test = int(input("房间数："))
y_pred = W[1]*x1_test + W[2]*x2_test + W[0]
print("预测价格：", round(y_pred,2),"万元")

预测房价
可视化：

from matplotlib import markers
import matplotlib.pyplot as plt
from mpl_toolkits.mplot3d import Axes3D
import numpy as np

x1 = x1 = np.array([137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00,
               106.69, 138.05, 53.75,  46.91,  68.00, 63.02, 81.26,  86.21])
x2 = np.array([3,2,2,3,1,2,3,2,2,3,1,1,1,1,2,2])
y = np.array([145.00, 110.00, 93.00,  116.00, 65.32, 104.00,118.00, 91.00,
              62.00,  133.00, 51.00,  45.00,  78.50, 69.65, 75.69,  95.30])

#绘制散点图
W = np.array([11.96729093, 0.53488599, 14.33150378])
y_pred = W[1]*x1 + W[2]*x2 + W[0]

fig = plt.figure(figsize=(8,6))
ax3d = Axes3D(fig)

ax3d.scatter(x1, x2, y, color="b",marker="*")

ax3d.set_xlabel('Area',color='r',fontsize=16)
ax3d.set_ylabel('Room',color='r',fontsize=16)
ax3d.set_zlabel('Price',color='r',fontsize=16)
ax3d.set_yticks([1,2,3])
ax3d.set_zlim3d(30,160)

plt.show()

#绘制平面图
X1,X2 = np.meshgrid(x1,x2)
Y_PRED = W[0] + W[1]*X1 + W[2]*X2

fig = plt.figure(2)
ax3d = Axes3D(fig)

ax3d.plot_surface(X1, X2, Y_PRED, cmap="coolwarm")

ax3d.set_xlabel('Area',color='r',fontsize=16)
ax3d.set_ylabel('Room',color='r',fontsize=16)
ax3d.set_zlabel('Price',color='r',fontsize=16)
ax3d.set_yticks([1,2,3])

plt.show()

#绘制散点图和线框图
plt.rcParams['font.sans-serif'] = ['SimHei']

fig = plt.figure(3)
ax3d = Axes3D(fig)

ax3d.scatter(x1,x2,y,color='b',marker='*',label="销售记录")
ax3d.scatter(x1,x2,y_pred,color='r',label="预测房价")
ax3d.plot_wireframe(X1,X2,Y_PRED,color='c',linewidth=0.5,label="拟合平面")


ax3d.set_xlabel('Area',color='r',fontsize=14)
ax3d.set_ylabel('Room',color='r',fontsize=14)
ax3d.set_zlabel('Price',color='r',fontsize=14)
ax3d.set_yticks([1,2,3])

plt.suptitle("商品房销售回归模型", fontsize=20)
plt.legend(loc="upper left")
plt.show()

在这里插入图片描述

总结

记一下学习笔记！

初学者___

关注

0
点赞
踩
2

收藏

觉得还不错? 一键收藏
3
评论
回归问题

学习笔记（1）一、解析法实现一元线性回归代码如下（示例）：# 解析法实现一元线性回归# 手动导入数据x = [137.97, 104.50, 100.00, 124.32, 79.20, 99.00, 124.00, 114.00, 106.69, 138.05, 53.75, 46.91, 68.00, 63.02, 81.26, 86.21]y = [145.00, 110.00, 93.00, 116.00, 65.32, 104.00,118.00, 91.00,
复制链接

扫一扫